get_max_similar_clustering: get_max_similar_clustering

In eytan-weinstein/scClustMatch: scClustMatch: A data visualization tool for interpreting the effects of reference choice on scRNAseq clustering results

get_max_similar_clustering

Description

Clusters the reads aligned to Reference Genome #2 iteratively at increasing increments of increment resolutions; starting at init_resolution and stopping only when the adjusted Rand index (ARI) similarity measure between the user's Reference Genome #1 clustering and the given Reference Genome #2 clustering has been globally maximized with an error of +/- increment resolutions. See the README for this package for more details regarding the mathematics of this algorithm.

Usage

get_max_similar_clustering(
  first_clustering,
  nearest_neighbors,
  dims,
  init_resolution,
  increment
)

Arguments

first_clustering	The S4 clustering class storing information related to the user's Reference Genome #1 clustering.
nearest_neighbors	The nearest neighbors graph construction computed from the alignment of reads to Reference Genome #2.
dims	The dimensions of reduction with which to cluster those reads.
init_resolution	The initial resolution at which to cluster those reads.
increment	The increment at which to cluster iteratively at increasing resolutions beginning with init_resolution.

Value

An S4 clustering class storing information related to that Reference Genome #2 clustering which is maximally similar to the user's first_clustering with an error of +/- increment resolutions.

eytan-weinstein/scClustMatch documentation built on May 3, 2022, 7:31 p.m.